Project Summary Metastasis?a defining feature of advanced cancer?often represents a transition from curable to incurable disease. The metastatic cascade consists of a series of severe obstacles that cancer cells must overcome, each one highly inefficient and apparently stochastic; we are presently unable to predict whether, when and where metastases will occur. We propose to apply an ecological lens to metastasis. Specifically, we will investigate the processes driving the increased metastatic potential of circulating tumor cell (CTC) clusters through a combination of mathematical modeling and in vivo quantitative experiments in a zebrafish model of melanoma. Melanoma, the most lethal of skin cancers, shows a particularly stark difference between the outcomes of patients with local versus metastatic disease: Patients with CTC clusters in their blood have worse clinical prognoses. Despite their importance, the mechanisms underlying CTC cluster formation, increased metastatic capacity, and potential for therapeutic targeting remain understudied?particularly in melanoma. We take advantage of the zebrafish model of metastatic melanoma, including the ZMEL1 cell line capable of transplantation into transparent Casper zebrafish, which provides a powerful system to quantitatively investigate the mechanisms behind increased metastatic potential of CTC clusters from an ecological perspective. Our three specific aims address how CTC clusters relate to metastatic fitness: (Aim 1) We hypothesize that the trade-off between group size and number?integral to ecological dispersal?is key in metastasis formation by CTC clusters; we will we will test this hypothesis with mathematical models to predict how the success of melanoma clusters varies with size, and we will confront those models with zebrafish data to quantify the metastatic fitness landscape of melanoma CTC clusters; we will then introduce genetic perturbations on hypothesized mechanisms of cellular cooperation within-clusters and elucidate the mechanisms underlying the shape of the cluster fitness landscape. (Aim 2) We hypothesize that high intra- cluster diversity promotes overall metastatic fitness despite the presence of some cells with lower individual fitness; we will test this hypothesis by engineering clusters with melanoma-specific forms of genetic heterogeneity; we will apply quantitative statistical analyses to compare high- and low-diversity clusters transplanted into zebrafish and evaluate the role of compositional heterogeneity in CTC cluster metastatic fitness using multi-level selection theory. (Aim 3) We hypothesize that microenvironmental gradients of diffusible substances determine the success of clusters of extravasated cells; we will test this hypothesis by investigating gradients in vivo, in vitro and in silico using an agent-based model with partial differential equations of reaction-diffusion. These aims, coupled with validation in mammalian models, will generate new insights into the dynamical processes underlying CTC cluster fitness towards the development of new diagnostic, prognostic and therapeutic strategies in melanoma and other cancers.